(var = hadsex)
age 14 = those who debuted before or at age 14 age 100 = those who have never debuted w/ opposite sex
go back and remind myself how “agefsex”
what proportion of all rels are marriages, cohabs, others at cross-section?
time between rel start and cohabitation
(marraiges who cohabitated premaritally)
mean of active relationships (where we assume exponential dist) plus estimates from survival analysis (using all reported relationships), exponential model fit
pavel recommends: “For weighted data the pseudo-populations size should be at least \(1 * sampleSize/smallestWeight\) (or \(3 * sampleSize/smallestWeight\) to be safe), or 1000 (whichever is larger).”
summary of NSFG weights:
summary(egodat$egos$weight)
## Min. 1st Qu. Median Mean 3rd Qu. Max.
## 38.7 2706.6 5199.6 8496.8 10138.2 85207.9
NSFG sample size: 43303
ppize for \(1 * sampleSize/smallestWeight\) = 1119
ppize for \(3 * sampleSize/smallestWeight\) = 3357
so round up to an even 3500 for estimation?
when ppop = 3500, you get actual ppop of ~500
and when ppop = 5000 you get actual pop of ~990 why?
raw data pulled from The National Vital Statistics Report Life Tables for 2017